On Monday the SemTech project had a face to face meeting in London to update on progress with the project and their survey of semantic technologies being used in education.
The day started with Thanassis Tiropanis giving an overview of the project today and in particular the survey site (see previous blog post) which has collated 40 semantic applications that can/are being used in teaching and learning. The team are now grappling with trying to make sense of the data collected. Some early findings, perhaps not surprisingly, show that there is most activity around information collection, publishing and data gathering. However there are some examples of more collaborative type activities being supported through semantic technologies. There is still time to contribute to the survey if you want to add any or your experiences.
I found the afternoon group discussions the most interesting part of the day. I chaired a group looking at the institutional perspective around using/adopting semantic technologies in respect of the following four questions:
1 what are the most important challenges in HE today?
2 how might semantic tech be part of the solution?
3 what are the current barriers to semantic technology adoption?
4 what areas of semantic technology require investing additional effort in?
As you would expect we had a fairly wide ranging discussion; but ultimately agreed that the key to getting some institutional traction would be to have some examples/use cases of how semantic technologies could help with key institutional concerns such as student retention. We came to the consensus that if data was more rigorously defined,categorized and normalized ie in RDF/triple stores, then it would be easier to query disparate data sources with added intelligence and so provide more tailored feedback/ early warning signs to teachers and administrators. However at the moment most institutions suffer from having numerous data empires who don’t see the need to communicate with each other and who don’t always have the most rigorous approach to data quality. Understanding data workflow within the institution is central to this. It will be interesting to see if any of the current JISC Curriculum Design projects decide to adopt a more semantic approach to workflow issues.
So in the answers we came up with were:
1 External influences eg HEFC, student retention, recruitment, course provisioning, research profile
2 Let us think of the questions we haven’t thought of yet.
3 Data empires, lack of knowledge, use cases, good examples in practice
4 Demonstrators to show value of adding semantic layer to existing data